“…Beyond this source surface, the field is assumed to be purely radial. The approximations implicit in this procedure have been discussed in detail elsewhere (Altschuler and Newkirk, 1969;Schatten, Wilcox, and Ness, 1969;Levine, 1978); they include the use of 27-day Carrington maps (which limit the temporal resolution), the neglect of possible coronal currents (which would distort the fields), and the use of a constant-radius source surface (which affects the size of the derived open field regions, i.e., coronal holes). Wang and Sheeley (1990a) used the same source-surface model with the slightly different radius R s = 2.5 Rp, which provides agreement with the observed sizes of the polar holes near sunspot minimum (Sheeley, Wang, and Harvey, 1989) and with the observed sector polarity of the interplanetary field (Hoeksema, 1984).…”